ROMJIST Volume 23, No. S, 2020, pp. S117-S129
Houssam BENBRAHIM, Hanaâ HACHIMI and Aouatif AMINE Deep Transfer Learning with Apache Spark to Detect COVID-19 in Chest X-ray Images
ABSTRACT: A chest X-ray test is one of the most important and recurrent medical imaging examinations. It is the first imaging technique that represents a significant role in the diagnosis of SARS-CoV-2 disease. Automatic classification of 2019-nCoV using X-ray images is a major request that can help doctors to make the best decisions. In this paper, we adopted, developed, and validated a Deep Transfer Learning (DTL) method using Convolutional Neural Network (CNN) based models InceptionV3 and ResNet50 with Apache Spark framework for the classification of COVID-19 in chest X-ray images collected from Kaggle repository. High accuracy was obtained by our model in the detection of COVID-19 X-ray images 99.01% by the pre-trained InceptionV3 model and 98.03% for the ResNet50 model.KEYWORDS: COVID-19, SARS-CoV-2, 2019-nCoV, chest X-ray images, deep transfer learning, convolutional neural network, CNN, apache spark, InceptionV3, ResNet50Read full text (pdf)